Interrupted Aortic Arch
Impact of Subaortic Stenosis on Management and Outcome
Abstract Interrupted aortic arch (IAA) is often related developmentally to subaortic obstruction (SAO). When severe, SAO must be addressed in surgical management of IAA. From 1990 to 1993, 25 neonates presented for initial surgical management of IAA complexes. Associated lesions were ventricular septal defect (VSD) with or without atrial septal defect (19 patients), truncus arteriosus (3 patients), tricuspid atresia with transposition of the great arteries (1 patient), aortic atresia with VSD (1 patient), and d-transposition of the great arteries with VSD (1 patient). Overall hospital mortality was 20% (five deaths). One death was related to sepsis and two to sudden hemodynamic decompensation (a 2-kg premature infant after arch repair and VSD closure and a neonate with IAA–truncus arteriosus after arch repair and truncus repair with aortic root replacement). Two deaths were related to low cardiac output in patients with severe subaortic narrowing (<3 mm by two-dimensional echocardiography), which was not addressed surgically. Of 10 additional patients judged preoperatively to have severe SAO, 1 underwent resection of the infundibular septum together with VSD closure and arch reconstruction, and 9 underwent a modification of Norwood’s operation with arch reconstruction and proximal pulmonary artery to aortic anastomosis (7 with systemic to pulmonary artery shunts and 2 with right ventricle to pulmonary artery outflow tract reconstruction). One patient died 2 months after surgery of staphylococcal sepsis. All 9 others were discharged well. Subaortic narrowing is a physiologically important element of IAA complexes. When SAO is severe, satisfactory initial palliation can be achieved by a modification of Norwood’s operation.
Interrupted aortic arch occurs in association with a wide variety of intracardiac malformations. When two ventricles are present, there is almost always an interventricular communication,1 and frequently there is some degree of obstruction or hypoplasia of the outflow tract of the subaortic ventricle.2 The use of prostaglandin E1 has made it routinely possible to resuscitate and support neonates with interrupted aortic arch complexes in anticipation of surgical therapy.3 Nonetheless, despite marked improvement in the outcome of surgical therapy for a broad spectrum of complex heart malformations in neonates, the operative mortality for repair of interrupted aortic arch complexes remains greater than that associated with many procedures that pose comparable if not more formidable technical challenges to the congenital heart surgeon.
It is generally recognized that hypoplasia of the subaortic outflow tract is a common accompaniment of interruption of the aortic arch. While on a physiological basis nonrelief of systemic ventricular outflow obstruction would be anticipated to impact on the physiology and potentially on survival after surgical management of interrupted aortic arch, controversy exists on what surgical techniques if any should be used to address systemic ventricular outflow tract obstruction and what parameters define a degree of subaortic obstruction that is likely to be poorly tolerated. Among existing clinical investigations, one of the largest and most comprehensive studies of outcome in patients with interrupted aortic arch complexes is the report by Jonas et al4 of the multi-institutional study undertaken by the Congenital Heart Surgeons Society. The analysis of outcomes among this group of 183 neonates with interrupted aortic arch and ventricular septal defect suggested that “procedural risk factors for death after repair were (1) repair without concomitant procedures in patients with other important levels of obstruction in the left heart–aortic complex, (2) a Damus-Kaye-Stansel anastomosis, and (3) subaortic myotomy/myectomy in the face of subaortic narrowing.” One is then left with the troublesome inferences that on the one hand failure to address the systemic ventricular outflow tract obstruction is associated with increased operative mortality, while on the other hand surgical modifications designed to relieve or bypass systemic ventricular outflow tract obstruction were also associated with increased operative mortality. While the Congenital Heart Surgeons Society study is unique because of the large size of the patient population, a potential limitation of that report is the incalculable contribution of interinstitutional variability in patient evaluation and management.
In an attempt to assess the importance of subaortic narrowing in patients with interrupted aortic arch complexes and determine the effectiveness of a surgical strategy that would be expected to minimize the influence of subaortic narrowing on postoperative physiology, we undertook this review of a single institution’s recent experience with surgical management of neonates with interrupted aortic arch.
All newborn infants with a diagnosis of interrupted aortic arch who were managed at the Children’s Hospital of Philadelphia between January 1990 and December 1993 constitute the study population. The population consists of 25 patients whose age at surgery ranged from 2 to 73 days. The latter was a premature infant maintained on ventilatory support and prostaglandin infusion at another institution until referral to the Children’s Hospital of Philadelphia for surgical management. Patient weight at the time of surgery ranged from 1.9 to 4.7 kg. In all instances, the diagnosis of interrupted aortic arch was made by two-dimensional echocardiography. The majority of patients did not undergo preoperative cardiac catheterization. Interrupted aortic arch was associated with normally related great arteries and single or multiple ventricular septal defects in 19 patients, with persistent truncus arteriosus communis in 3, tricuspid atresia and d-transposition of the great arteries in 1, aortic atresia and malalignment-type ventricular septal defect in 1, and d-transposition of the great arteries and malalignment-type ventricular septal defect in 1.
Assessment of the anatomy including echocardiographic measurement of the subaortic region of the left ventricular outflow tract in patients with normally related great arteries was made at the time of diagnosis by one of five attending staff cardiologists/echocardiographers. As part of this retrospective study, all available echocardiographic tapes were reviewed, and measurements of the subaortic region of the left ventricular outflow tract were made by a single observer (A.J.C.).
All operative procedures were performed by two surgeons using one general scheme of operative management.
The primary source of follow-up information is the hospital medical record. The secondary source of follow-up information is the records of the referring cardiologists. None of the 25 patients have been lost to follow-up.
Measurements of the subaortic region of the left ventricular outflow tract by one observer are summarized in the Table⇓ for the 19 patients with interrupted aortic arch and ventricular septal defect. Also listed in the Table⇓ are the anatomic type of the ventricular septal defect and the nature of the aortic valve (bicuspid or otherwise) when known. While the vast majority of measurements of the subaortic region of the left ventricular outflow tract are clustered in the 3 to 5 mm range, there is for most patients some degree of variability between measurements made using different echocardiographic windows. Similarly, there were at times larger differences between these measurements made retrospectively by one observer and those that were made by a variety of echocardiographers at the time of initial assessment.
Of the 19 patients with interrupted aortic arch and ventricular septal defect, 11 underwent ventricular septal defect closure and anastomosis of the ascending aorta to the thoracic aorta with augmentation of the ascending aorta, aortic arch, and thoracic aorta by use of a gusset of cryopreserved pulmonary artery homograft (Fig 1⇓). In 1 of these 11 patients, the conal septum was resected to enlarge the subaortic portion of the left ventricular outflow tract. One patient with interrupted aortic arch with ventricular septal defect underwent ventricular septal defect closure and simple primary anastomosis of the postductal thoracic aorta to the ascending aorta. Seven patients with interrupted aortic arch and ventricular septal defect who were believed to have severe subaortic stenosis underwent a modified Norwood procedure wherein the transected proximal main pulmonary artery was associated with the ascending aorta, and the aortic arch was repaired by anastomosis of the thoracic aorta to the ascending aorta with cryopreserved pulmonary artery homograft augmentation of the aortic arch (Fig 2⇓). In 5 of these patients, pulmonary blood flow was provided by a systemic to pulmonary artery shunt, as in the Norwood procedure as described for hypoplastic left heart syndrome.5 In the remaining 2 patients, an interventricular baffle was used to direct left ventricular outflow through the ventricular septal defect to the pulmonic valve, and the branch pulmonary arteries were associated with the right ventricular outflow tract, resulting in a one-stage biventricular repair. Two additional patients, one with interrupted aortic arch, tricuspid atresia, and d-transposition of the great arteries and one with interrupted aortic arch, aortic atresia, and malalignment-type ventricular septal defect, underwent a Norwood stage I procedure. The patient with interrupted aortic arch, d-transposition of the great arteries, and malalignment-type ventricular septal defect underwent ventricular septal defect closure, arterial switch, aortic arch anastomosis with pulmonary artery homograft gusset augmentation, and transannular patch reconstruction of the right ventricular outflow tract. Two of the three patients with truncus arteriosus and interrupted aortic arch underwent arch anastomosis with homograft augmentation, ventricular septal defect closure, and truncus repair with right ventricular outflow tract reconstruction. One of the two also underwent aortic root replacement with a cryopreserved aortic homograft and coronary reimplantation. The third patient with truncus arteriosus and interrupted aortic arch also had severe truncal valve dysplasia and regurgitation and underwent orthotopic cardiac and aortic arch transplantation.
Outcome of the Initial Procedure
There were five hospital deaths, for a hospital mortality of 20%. The patient who had undergone repair of truncus arteriosus, aortic arch repair, and aortic root replacement experienced sudden hemodynamic decompensation on the third postoperative day and died. One of the two patients who underwent a Norwood-type one-stage biventricular repair died as a result of staphylococcal sepsis 2 months after surgery. Among the 13 patients who underwent ventricular septal defect closure and aortic arch repair with or without homograft augmentation, there were three hospital deaths. The first was a 2-kg, premature infant who experienced sudden hemodynamic decompensation on the first postoperative day and died. The other two patients who died remained ventilator dependent with poor cardiac output and died as a result of multiple organ dysfunction on postoperative days 10 and 23, respectively. Analysis of survival in relation to method of initial repair reveals that among patients with interrupted aortic arch and ventricular septal defect who underwent arch repair and ventricular septal defect closure, survival was 77% at 1 month and remained at 77% at 6 and 12 months. For patients who underwent a Norwood-type repair including association of the proximal main pulmonary artery with the ascending aorta, survival was 100% at 1 month and 89% at 6 and 12 months.
The frequency and severity of hypoplasia or stenosis of the subaortic region of the left ventricular outflow tract in association with interrupted aortic arch have been widely recognized and are well documented.2 6 7 It makes sense on a physiological basis to hypothesize that failure to relieve systemic ventricular outflow obstruction at levels proximal to that of the corrected arch interruption could have an important impact on postoperative physiology and survival. We can only speculate whether failure to address physiologically important subaortic stenosis was a factor contributing to the death of 3 of the 13 patients in this series who underwent ventricular septal defect closure and arch repair. One of the three had the smallest subaortic stenotic region of the entire patient population by retrospective echocardiographic measurement.
There remains considerable controversy on the importance of obstruction of the subaortic region in patients with interrupted aortic arch and the most appropriate surgical strategy to deal with it when present.8 9 10 In 1988, Sell and associates11 in Boston reported an improvement in outcome associated with primary one-stage repair of interrupted aortic arch and ventricular septal defect by ventricular septal defect closure and direct aortic anastomosis. They predicted and observed a high incidence of reintervention for left ventricular outflow tract obstruction among survivors but did not suggest any specific alteration of the initial surgical strategy for patients with more than usually severe degrees of subaortic narrowing. Ilbawi et al10 and Bove et al9 independently reported encouraging results in small groups of patients who were believed to have severe subaortic narrowing and were managed by myotomy and myectomy of the left ventricular outflow tract in association with ventricular septal defect closure and arch repair. Yet in the Congenital Heart Surgeons Society study, mortality after an initial procedure including myotomy/myectomy was 47%. A different strategy incorporating an end-to-side proximal main pulmonary artery to aortic anastomosis (referred to in the Congenital Heart Surgeons Society report as a Damus-Kaye-Stansel anastomosis) was reportedly used in that series in a total of 11 patients at a variety of participating centers, with uniformly poor results (mortality, 91%). It is discouraging that in that careful analysis of a large group of patients by the members of the Congenital Heart Surgeons Society, ventricular septal defect closure and arch repair without a concomitant procedure directed at relief of coexisting left heart outflow obstruction was an incremental risk factor for death after repair. Yet at the same time, some strategies directed at subaortic obstruction including myotomy/myectomy and Damus-Kaye-Stansel anastomosis were associated with high rates of operative mortality.
The Norwood operation, which provides rational and satisfactory initial palliation for hypoplastic left heart syndrome, is equally applicable to a variety of other heart malformations characterized by systemic ventricular outflow tract obstruction and ductal dependency of the systemic circulation.12 Interrupted aortic arch complexes fall into that category of malformations whose physiology is dominated by ductal dependency of the systemic circulation and is usually characterized by a degree of systemic ventricular outflow tract obstruction. Thus, the Norwood operation should be associated with predictably satisfactory results in the initial management of patients with interrupted aortic arch complexes. Our practice in the last several years has been to use this operative strategy when the available preoperative and intraoperative data suggest that a degree of systemic ventricular outflow tract obstruction is present that could impact negatively on survival after a more conventional reparative procedure. Such is the case when the preoperative echocardiographic assessment suggests a severe degree of narrowing or hypoplasia of the subaortic region of the left ventricular outflow tract. In such circumstances, the Norwood operation is performed exactly as for hypoplastic left heart syndrome, with the only additional feature being association of the ascending aorta with the descending thoracic aorta together with homograft arch augmentation, as shown in Fig 1⇑. Alternatively, as in two patients in this series, an interventricular baffle may be used to direct left ventricular outflow through the ventricular septal defect to the pulmonic valve which has been associated with the aortic arch elements and a conduit used to associate the branch pulmonary arteries with the right ventricle, thus accomplishing a one-stage biventricular repair. For patients who undergo the more conventional Norwood operation with systemic to pulmonary artery shunt, biventricular repair may later be accomplished, as it is in instances of aortic atresia with malalignment-type ventricular septal defect. Alternatively, these patients may be managed by staged reconstruction culminating in a Fontan operation if mandated, as for example by hypoplasia of other left heart structures, including the mitral valve.
In instances in which assessment of the anatomy and morphology of the left ventricular outflow tract is not predictive of a severe degree of left ventricular outflow obstruction, it is our usual practice to accomplish ventricular septal defect closure and association of ascending and descending aortic elements with aortic arch augmentation by use of a gusset of cryopreserved arterial homograft. Minimizing the likelihood of any degree of aortic obstruction from the level of the sinotubular junction through the arch and down to the thoracic aorta may in fact, as suggested by Jonas et al,4 enhance the likelihood of patient survival in the face of a mild to moderate degree of subaortic narrowing. When preoperative studies have suggested the presence of a severe degree of subaortic narrowing, we have been encouraged by the satisfactory results of initial palliation that can be achieved by modification of the Norwood operation.
- Copyright © 1995 by American Heart Association
Ho SY, Wilcox BR, Anderson RH, Lincoln JCR. Interrupted aortic arch: anatomical features of surgical significance. J Thorac Cardiovasc Surg. 1983;31:199-205.
Leoni F, Huhta JC, Douglas J, et al. Effect of prostaglandin on early surgical mortality in obstructive lesions of the systemic circulation. Br Heart J. 1984;52:654-659.
Jacobs ML, Norwood WI. Hypoplastic left heart syndrome. In: Pediatric Cardiac Surgery: Current Issues. Butterworth-Heinemann; 1992:182-192.